Creator Mode and Game Mode

*This page was updated on 8/17. A subsequent article with new information has been posted.

Due to the difference in memory latency between the two pairs of memory channels, AMD is implementing a ‘mode’ strategy for users to select depending on their workflow. The two modes are called Creator Mode (default), and Game Mode, and control two switches in order to adjust the performance of the system.

The two switches are:

  • Legacy Compatibility Mode, on or off (off by default)
  • Memory Mode: UMA vs NUMA (UMA by default)

The first switch disables the cores in one fo the silicon dies, but retains access to the DRAM channels and PCIe lanes. When the LCM switch is off, each core can handle two threads and the 16-core chip now has a total of 32 threads. When enabled, the system cuts half the cores, leaving 8 cores and 16 threads. This switch is primarily for compatibility purposes, as certain games (like DiRT) cannot work with more than 20 threads in a system. By reducing the total number of threads, these programs will be able to run. Turning the cores in one die off also alleviates some potential pressure in the core microarchitecture for cross communication.

The second switch, Memory Mode, puts the system into a unified memory architecture (UMA) or a non-unified memory architecture (NUMA) mode. Under the default setting, unified, the memory and CPU cores are seen as one massive block to the system, with maximum bandwidth and an average latency between the two. This makes it simple for code to understand, although the actual latency for a single instruction will be a good +20% faster or slower than the average, depending on which memory bank it is coming from.

NUMA still gives the system the full memory, but splits the memory and cores into into two NUMA banks depending on which pair of memory channels is nearest the core that needs the memory. The system will keep the data for a core as near to it as possible, giving the lowest latency. For a single core, that means it will fill up the memory nearest to it first at half the total bandwidth but a low latency, then the other half of the memory at the same half bandwidth at a higher latency. This mode is designed for latency sensitive workloads that rely on the lower latency removing a bottleneck in the workflow. For some code this matters, as well as some games – low latency can affect averages or 99th percentiles for game benchmarks.

The confusing thing about this switch is that AMD is calling it ‘Memory Access Mode’ in their documents, and labeling the two options as Local and Distributed. This is easier to understand than the SMT switch, in that the Local setting focuses on the latency local to the core (NUMA), and the Distributed setting focuses on the bandwidth to the core (UMA), with Distributed being default.

  • When Memory Access Mode is Local, NUMA is enabled (Latency)
  • When Memory Access Mode is Distributed, UMA is enabled (Bandwidth, default)

So with that in mind, there are four ways to arrange these two switches. AMD has given two of these configurations specific names to help users depending on how they use their system: Creator Mode is designed to give as many threads as possible and as much memory bandwidth as possible. Game Mode is designed to optimize for latency and compatibility, to drive game frame rates.

AMD Threadripper Options
  Words That Make Sense   Marketing Spiel
Ryzen
Master
Profile
Two Dies or
One Die
Memory
Mode
  Legacy
Compatibility
Mode
Memory
Access
Mode
Creator Mode Two UMA   Off Distributed
- Two NUMA   Off Local
- One UMA   On Distributed
Game Mode One NUMA   On Local

There are two ways to select these modes, although this is also a confusing element to this situation.

The way I would normally adjust these settings is through the BIOS, however the BIOS settings do not explicitly state ‘Creator Mode’ and ‘Game Mode’. They should give immediate access for the Memory Mode, where ASUS has used the Memory Access naming for Local and Distributed, not NUMA and UMA.  For the Legacy Compatibility Mode, users will have to dive several screens down into the Zen options and manually switch off eight of the cores, if the setting is going to end up being visible to the user. This makes Ryzen Master the easiest way to implement Game Mode.

While we were testing Threadripper, AMD updated Ryzen Master several times to account for the latest updates, so chances are that by the time you are reading this, things might have changed again. But the crux is that Creator Mode and Game Mode are not separate settings here either. Instead, AMD is labelling these as ‘profiles’. Users can select the Creator Mode profile or the Game Mode profile, and within those profiles, the two switches mentioned above (labelled as Legacy Compatibility Mode and Memory Access Mode) will be switched as required.

Cache Performance

As an academic exercise, Creator Mode and Game Mode make sense depending on the workflow. If you don’t need the threads and want the latency bump, Game Mode is for you. The perhaps odd thing about this is that Threadripper is aimed at highly threaded workloads more than gaming, and so losing half the threads in Game Mode might actually be a detriment to a workstation implementation.  That being said, users can leave SMT on and still change the memory access mode on its own, although AMD is really focusing more on the Creator and Game mode specifically.

For this review, we tested both Creator (default) and Game modes on the 16-core Threadripper 1950X. As an academic exercise we looked into memory latency in both modes, as well as at higher DRAM frequencies. These latency numbers take the results for the core selected (we chose core 2 in each case) and then stride through to hit L1, L2, L3 and main memory. For UMA systems like in Creator Mode, main memory will be an average between the near and far memory results. We’ve also added in here a Ryzen 5 1600X as an example of a single Zeppelin die, and a 6950X Broadwell for comparison. All CPUs were run at DDR4-2400, which is the maximum supported at two DIMMs per channel. 

For the 1950X in the two modes, the results are essentially equal until we hit 8MB, which is the L3 cache limit per CCX. After this, the core bounces out to main memory, where the Game mode sits around 79ns while the Creator mode is at 108 ns. By comparison the Ryzen 5 1600X seems to have a lower latency at 8MB (20ns vs 41 ns), and then sits between the Creator and Game modes at 87 ns. It would appear that the bigger downside of Creator mode in this way is the fact that main memory accesses are much slower than normal Ryzen or in Game mode.

If we crank up the DRAM frequency to DDR4-3200 for the Threadripper 1950X, the numbers change a fair bit:


Click for larger image

Up until the 8MB boundary where L3 hits main memory, everything is pretty much equal. At 8MB however, the latency at DDR4-2400 is 41ns compared to 18ns at DDR4-3200. Then out into full main memory sees a pattern: Creator mode at DDR4-3200 is close to Game Mode at DDR4-2400 (87ns vs 79ns), but taking Game mode to DDR4-3200 drops the latency down to 65ns.

Another element we tested while in Game Mode was the latency for near memory and far memory as seen from a single core. Remember this slide from AMD’s deck?

In our testing, we achieved the following:

  • At DDR4-2400, 79ns near memory and 136ns far memory (108ns average)
  • At DDR4-3200, 65ns near memory and 108ns far memory (87ns average)

Those average numbers are what we get for Creator mode by default, indicating that the UMA mode in Creator mode will just use memory at random between the two.

Silicon, Glue, & NUMA Too Test Bed and Setup
Comments Locked

347 Comments

View All Comments

  • Pekish79 - Friday, August 11, 2017 - link

    Vraybench 1.0.5
  • SanX - Friday, August 11, 2017 - link

    *** AMD, make 2-chip mobos for upcoming multicore wars, you will double your profit from this at no cost for you +++
  • vicbee - Friday, August 11, 2017 - link

    Off subject: Having just read the article about nVidia's meteoric rise in profits, some of which directly attributed to high end "gamers" video cards purchased expressly for coin mining, I wonder if it and AMD are going to manufacture CPU's and GPU's specifically for that purpose and how that will affect the price of said parts...
  • Avro Arrow - Friday, August 11, 2017 - link

    Hi Ian, thanks for doing this article. It's important to see all possible outcomes because in the real world, anything is possible. I do have one question that has be puzzled. Why do you say that Threadripper only has 64 PCI-Express 3.0 lanes when it's been reported several times by everyone, including official AMD releases (and also including by you) that it has 64? I thought it might be just a typo but you state it in several places and in all of your specs. This is not a new thing so is there something about Threadripper that we don't know?
  • HotJob - Friday, August 11, 2017 - link

    Could someone explain to me what a "2P" system is from the competition section of the article?
  • coolhardware - Saturday, August 12, 2017 - link

    "2P" system = two processor system, i.e. a system with two physical CPU sockets and two CPUs installed.

    In the past a 2P (or 4P) system was really handy to get more cores especially back when 1 core, 2 core, and eventually 4 core CPUs were high end. In the consumer realm, way back, the Pentium II was the first 2P system I ever built and people even did it with Celerons as well:
    http://www.cpu-central.com/dualceleron/
    the Opterons were also fun for dual or quad processor systems including some SFF options like the ZMAX-DP socket 940 system.
    https://www.newegg.com/Product/Product.aspx?Item=N...

    Now fast forward with ThreadRipper already available at Amazon and NewEgg
    http://amzn.to/2wDqgWw (URL shortened)
    https://www.newegg.com/Product/Product.aspx?Item=N...
    I do not think I will ever be building a 2P or 4P system again!!!

    :-)
  • rvborgh - Friday, August 11, 2017 - link

    hi Ian,

    i think the Cinebench 11.5 benchmarks are incorrect for both ThreadRippers. ThreadRipper is almost equivalent to my Quad Opteron (48 core) system which scores 3229cb on R15... and 39.04 on Cinebench 11.5. if i downclock all cores to approximately 2.9 GHz i end up with around 3000cb in R15 and in the 36 range point range for 11.5.

    The fact that you are only scoring in the 18 range makes me wonder if you had the Threadripper set in some mode where it was only using 8 out of the 16 cores. Can you verify this... please? Thanks :) i would think you should see scores in the 36 range with 11.5.

    Other than this minor detail... great article.

    PS: i've had the same issues with software not liking NUMA on my quad opteron system as well... Cinebench especially does not like it.
  • Tchamber - Saturday, August 12, 2017 - link

    Hi, Ian. Thanks for the review. As usual it was in depth and informative. I'm in the middle of building a 1700x system now based on your review. I wanted to say you handle all the nay-Sayers, gloomy Gusses and negative Nacies with aplomb! I think most people's own slant colors how they see your reviews. I appreciate the consistency of what you do here. I took a look over at Ars, and they could be called AMD shills for all the positive things they say... Keep it up!
  • Tchamber - Saturday, August 12, 2017 - link

    P.S.
    I loved your Kessel Run reference, it tied in nicely with your Yoda quote.
  • B3an - Saturday, August 12, 2017 - link

    Too many plebs complaining about a lack of 3D rendering benches. The fact is a 16 core CPU is still much slower than GPU's at rendering. I'll be getting a 1950X but it wont even be used for rendering when i know for a fact that my two GPUs will still be much faster with things like Blender. Even a single high-end GPU will still easily beat the 1950X at these tasks.

    Seems like immature moron fanboys are crying over this stuff because they just want to see AMD at the top of the charts.

Log in

Don't have an account? Sign up now